submission_id: gryphe-tiamat-8b-1-2-lla_6673_v1
developer_uid: Gryphe
status: inactive
model_repo: Gryphe/Tiamat-8b-1.2-Llama-3-DPO
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
reward_formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
timestamp: 2024-05-08T13:04:04+00:00
model_name: gryphe-tiamat-8b-1-2-lla_6673_v1
model_eval_status: success
double_thumbs_up: 106
thumbs_up: 157
thumbs_down: 87
num_battles: 10176
num_wins: 4917
celo_rating: 1154.89
entertaining: 6.96
stay_in_character: 8.71
user_preference: 7.24
safety_score: 0.86
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030523392.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: gryphe-tiamat-8b-1-2-lla_6673_v1
double_thumbs_up_ratio: 0.3028571428571429
feedback_count: 350
ineligible_reason: None
language_model: Gryphe/Tiamat-8b-1.2-Llama-3-DPO
model_score: 7.636666666666668
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
single_thumbs_up_ratio: 0.44857142857142857
thumbs_down_ratio: 0.24857142857142858
thumbs_up_ratio: 0.7514285714285714
us_pacific_date: 2024-05-08
win_ratio: 0.4831957547169811
Resubmit model
Running pipeline stage MKMLizer
Starting job with name gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer
Waiting for job on gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer to finish
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ _____ __ __ ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ /___/ ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ Version: 0.8.10 ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ The license key for the current software has been verified as ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ belonging to: ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ Chai Research Corp. ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ║ ║
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Downloaded to shared memory in 29.829s
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: quantizing model to /dev/shm/model_cache
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 44%|████▎ | 127/291 [00:00<00:01, 127.08it/s] Loading 0: 89%|████████▊ | 258/291 [00:01<00:00, 129.38it/s] Loading 0: 99%|█████████▊| 287/291 [00:07<00:00, 29.46it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: quantized model in 17.753s
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Processed model Gryphe/Tiamat-8b-1.2-Llama-3-DPO in 48.662s
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: creating bucket guanaco-mkml-models
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/gryphe-tiamat-8b-1-2-lla-6673-v1
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/gryphe-tiamat-8b-1-2-lla-6673-v1/config.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/gryphe-tiamat-8b-1-2-lla-6673-v1/tokenizer_config.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/gryphe-tiamat-8b-1-2-lla-6673-v1/special_tokens_map.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/gryphe-tiamat-8b-1-2-lla-6673-v1/tokenizer.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: warnings.warn(
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: warnings.warn(
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:468: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: warnings.warn(
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: return self.fget.__get__(instance, owner)()
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Saving duration: 0.242s
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.777s
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: creating bucket guanaco-reward-models
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/gryphe-tiamat-8b-1-2-lla-6673-v1_reward
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/gryphe-tiamat-8b-1-2-lla-6673-v1_reward/config.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/gryphe-tiamat-8b-1-2-lla-6673-v1_reward/tokenizer_config.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/gryphe-tiamat-8b-1-2-lla-6673-v1_reward/merges.txt
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/gryphe-tiamat-8b-1-2-lla-6673-v1_reward/special_tokens_map.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/gryphe-tiamat-8b-1-2-lla-6673-v1_reward/vocab.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/gryphe-tiamat-8b-1-2-lla-6673-v1_reward/tokenizer.json
gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/gryphe-tiamat-8b-1-2-lla-6673-v1_reward/reward.tensors
Job gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer completed after 73.1s with status: succeeded
Stopping job with name gryphe-tiamat-8b-1-2-lla-6673-v1-mkmlizer
Pipeline stage MKMLizer completed in 78.50s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.14s
Running pipeline stage ISVCDeployer
Creating inference service gryphe-tiamat-8b-1-2-lla-6673-v1
Waiting for inference service gryphe-tiamat-8b-1-2-lla-6673-v1 to be ready
Inference service gryphe-tiamat-8b-1-2-lla-6673-v1 ready after 40.23633575439453s
Pipeline stage ISVCDeployer completed in 48.31s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.7441723346710205s
Received healthy response to inference request in 1.102647066116333s
Received healthy response to inference request in 1.106015920639038s
Received healthy response to inference request in 0.9750945568084717s
Received healthy response to inference request in 1.0901679992675781s
5 requests
0 failed requests
5th percentile: 0.998109245300293
10th percentile: 1.0211239337921143
20th percentile: 1.067153310775757
30th percentile: 1.0926638126373291
40th percentile: 1.097655439376831
50th percentile: 1.102647066116333
60th percentile: 1.103994607925415
70th percentile: 1.105342149734497
80th percentile: 1.2336472034454347
90th percentile: 1.4889097690582276
95th percentile: 1.616541051864624
99th percentile: 1.7186460781097412
mean time: 1.2036195755004884
Pipeline stage StressChecker completed in 6.71s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.04s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.04s
M-Eval Dataset for topic stay_in_character is loaded
gryphe-tiamat-8b-1-2-lla_6673_v1 status is now deployed due to DeploymentManager action
gryphe-tiamat-8b-1-2-lla_6673_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics